CN104575509A - Voice enhancement processing method and device - Google Patents
Voice enhancement processing method and device Download PDFInfo
- Publication number
- CN104575509A CN104575509A CN201410834628.1A CN201410834628A CN104575509A CN 104575509 A CN104575509 A CN 104575509A CN 201410834628 A CN201410834628 A CN 201410834628A CN 104575509 A CN104575509 A CN 104575509A
- Authority
- CN
- China
- Prior art keywords
- cement
- voice
- speech enhan
- terminal
- voice enhancement
- Prior art date
Links
- 238000003672 processing method Methods 0.000 title abstract 2
- 238000000034 methods Methods 0.000 claims abstract description 54
- 230000000875 corresponding Effects 0.000 claims abstract description 34
- 239000004568 cements Substances 0.000 claims description 155
- 230000000576 supplementary Effects 0.000 claims description 28
- 238000010586 diagrams Methods 0.000 description 6
- 280001015926 ABBYY companies 0.000 description 5
- 230000002708 enhancing Effects 0.000 description 5
- 230000001537 neural Effects 0.000 description 4
- 238000004458 analytical methods Methods 0.000 description 3
- 238000007664 blowing Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000000873 masking Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reactions Methods 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering processes Methods 0.000 description 1
- 230000002452 interceptive Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000003786 synthesis reactions Methods 0.000 description 1
- 230000002194 synthesizing Effects 0.000 description 1
Abstract
Description
Technical field
The present invention relates to Internet technical field, particularly relate to a kind of speech enhan-cement disposal route and device.
Background technology
Along with the acoustic enviroment of Intelligent hardware becomes increasingly complex, speech recognition for Intelligent hardware also more has challenge, time distant from microphone when a user speaks, Intelligent hardware likely can not identify the phonetic entry of user, therefore needs to carry out noise reduction and speech enhan-cement process to the voice of input.Prior art by arranging speech enhan-cement chip or carrying out speech enhan-cement by the voice of central processing unit (CPU) to input of Intelligent hardware in Intelligent hardware, if adopt the voice of speech enhan-cement chip to input to carry out speech enhan-cement process, when to speech enhan-cement quality requirements height, need to choose can be suitable for the high speech enhan-cement chip of computation complexity to promote speech enhan-cement quality, thus the hardware cost of terminal device can be improved, if adopt the voice of CPU to input to carry out speech enhan-cement, then can take and consume the local a large amount of computational resource of terminal device.
Summary of the invention
In view of this, the invention provides a kind of speech enhan-cement process disposal route and device, save hardware cost and the computational resource of terminal device further.
According to the first aspect of this method embodiment, provide a kind of speech enhan-cement disposal route, application on the server, comprising:
Obtain the voice messaging from terminal device, in described voice messaging, carry speech enhan-cement supplementary;
If judge to know that described voice messaging needs to carry out speech enhan-cement process according to the device identification of described terminal device, then according to described speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm;
Described voice enhancement algorithm according to obtaining carries out speech enhan-cement process to described voice messaging.
According to the second aspect of this method embodiment, provide a kind of speech enhan-cement treating apparatus, application on the server, comprising:
First acquisition module, for obtaining the voice messaging from terminal device, carries speech enhan-cement supplementary in described voice messaging;
Second acquisition module, if know that the described voice messaging that described first acquisition module obtains needs to carry out speech enhan-cement process for judging according to the device identification of described terminal device, then according to described speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm;
Speech enhan-cement module, carries out speech enhan-cement process for the described voice enhancement algorithm obtained according to described second acquisition module to described voice messaging.
From above technical scheme, the present invention judges to know that voice messaging needs to carry out speech enhan-cement process to according to the device identification of terminal device, according to speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm, owing to can be suitable on the server, therefore avoid and carry out speech enhan-cement on the terminal device, thus alleviate the computation burden of terminal device; Due to corresponding voice enhancement algorithm can be adopted to the voice messaging of terminal device, avoid unnecessary voice enhancement algorithm and enhancing process is carried out to voice messaging, make that speech enhan-cement is carried out to voice messaging and have more specific aim, thus the computation complexity of server when carrying out speech enhan-cement can be reduced substantially, improve the quality of speech enhan-cement, and then make follow-up speech recognition more accurate.
Should be understood that, it is only exemplary and explanatory that above general description and details hereinafter describe, and can not limit the embodiment of the present invention.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of speech enhan-cement disposal route in one embodiment of the present invention;
Fig. 2 is the process flow diagram of speech enhan-cement disposal route in another embodiment of the present invention;
Fig. 3 is the process flow diagram of speech enhan-cement disposal route in another way of example of the present invention;
Fig. 4 is the structural drawing of speech enhan-cement server in one embodiment of the present invention;
Fig. 5 is the system construction drawing of speech enhan-cement process in one embodiment of the present invention;
Fig. 6 is the building-block of logic of speech enhan-cement treating apparatus in one embodiment of the present invention;
Fig. 7 is the building-block of logic of speech enhan-cement treating apparatus in another embodiment of the present invention.
Embodiment
Here will be described exemplary embodiment in detail, its sample table shows in the accompanying drawings.When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or analogous key element.Embodiment described in following exemplary embodiment does not represent all embodiments consistent with the application.On the contrary, they only with as in appended claims describe in detail, the example of apparatus and method that some aspects of the application are consistent.
Only for describing the object of specific embodiment at term used in this application, and not intended to be limiting the application." one ", " described " and " being somebody's turn to do " of the singulative used in the application and appended claims is also intended to comprise most form, unless context clearly represents other implications.It is also understood that term "and/or" used herein refer to and comprise one or more project of listing be associated any or all may combine.
Term first, second, third, etc. may be adopted although should be appreciated that to describe various information in the application, these information should not be limited to these terms.These terms are only used for the information of same type to be distinguished from each other out.Such as, when not departing from the application's scope, the first information also can be called as the second information, and similarly, the second information also can be called as the first information.Depend on linguistic context, word as used in this " if " can be construed as into " ... time " or " when ... time " or " in response to determining ".
The application by server according to the voice enhancement algorithm of speech enhan-cement supplementary to the voice messaging determination speech enhan-cement of the terminal device got, and by corresponding voice enhancement algorithm, speech enhan-cement process is carried out to voice messaging, therefore avoid and carry out speech enhan-cement on the terminal device, thus alleviate the computation burden of terminal device; Due to corresponding voice enhancement algorithm can be adopted to the voice messaging of terminal device, thus can adopt and have more voice enhancement algorithm targetedly speech enhan-cement process is carried out to the voice messaging of terminal device, server is avoided to adopt the high voice enhancement algorithm of computation complexity to carry out unnecessary speech enhan-cement process to the voice messaging of terminal device, reduce server computation complexity when carrying out speech enhan-cement process substantially, improve the quality of speech enhan-cement, and then make follow-up speech recognition more accurate.For being further described the application, provide the following example.
Please refer to Fig. 1, Fig. 1 is the process flow diagram of speech enhan-cement disposal route in one embodiment of the present invention, can apply on the server, terminal device in the embodiment of the present invention can comprise: the various equipment with speech voice input function such as in-car TV, intelligent remote controller, smart mobile phone, panel computer, comprise the steps:
Step 101, obtains the voice messaging from terminal device, wherein, carries speech enhan-cement supplementary in voice messaging.
In one embodiment, by the microphones capture of terminal device to analog voice, after terminal device carries out analog to digital conversion and compress speech to analog voice, the voice messaging described in the embodiment of the present invention can be formed.
Step 102, if judge to know that voice messaging needs to carry out speech enhan-cement process according to the device identification of terminal device, then according to speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm.
Because speech enhan-cement not only relates to voice signal digital processing, also relate to Auditory Perception and the phonetics category of people, add the difference of environment residing for terminal device, noise source also can be different, thus voice enhancement algorithm and the environmental correclation residing for terminal device, in addition, due to the difference of the current duty of terminal device, terminal device also can be different by the analog voice that microphones capture arrives, such as, when terminal device is in hands-free mode and map mode, microphone can be easier to capture extraneous noise, therefore work state information and ambient parameter information can be sent to server in the mode of speech enhan-cement supplementary by the embodiment of the present invention, server is determined in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm by speech enhan-cement supplementary, thus can get and have more voice enhancement algorithm thus speech enhan-cement is carried out to voice messaging targetedly.
Step 103, the voice enhancement algorithm according to obtaining carries out speech enhan-cement process to voice messaging.
In one embodiment, such as, terminal device is in hands-free mode or map mode, for the voice messaging of terminal device being in hands-free mode and map mode, the voice enhancement algorithm that computation complexity is higher can be adopted to carry out speech enhan-cement, and for the terminal device under normal mode, the voice enhancement algorithm that computation complexity is lower can be adopted to carry out speech enhan-cement, making speech enhan-cement implementation procedure have more specific aim thus, the unnecessary computation burden of server can be reduced when guaranteeing speech enhan-cement quality.In another embodiment, terminal device is in (noise source is based on the noisy sound of people) in market, or, terminal device is in (noise source is based on the sound of blowing a whistle of vehicle) on road, or, terminal device is in classroom (essentially no noise), in such a case, if terminal device is in market, the voice enhancement algorithm that can adopt the noisy sound (can be identified by frequency) eliminating people carries out speech enhan-cement to the voice messaging of terminal device, if terminal device is positioned on road, the voice enhancement algorithm that can adopt the sound of blowing a whistle eliminating vehicle carries out speech enhan-cement to the voice messaging of terminal device, if terminal device is in classroom, better simply common voice enhancement algorithm can be adopted to carry out speech enhan-cement to the voice messaging of terminal device, make speech enhan-cement process adopt thus and have more voice enhancement algorithm targetedly.
As can be seen from step 101-step 103, the present invention judges to know that voice messaging needs to carry out speech enhan-cement process according to the device identification of terminal device, according to speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm, owing to can be suitable on the server, therefore avoid and carry out speech enhan-cement on the terminal device, thus alleviate the computation burden of terminal device; Due to corresponding voice enhancement algorithm can be adopted to the voice messaging of terminal device, avoid unnecessary voice enhancement algorithm and enhancing process is carried out to the voice messaging of terminal device, thus the computation complexity of server when carrying out speech enhan-cement can be reduced substantially, improve the quality of speech enhan-cement, and then make follow-up speech recognition more accurate.
Refer to Fig. 2, Fig. 2 is the process flow diagram of speech enhan-cement disposal route in another embodiment of the present invention, the present embodiment can be applied on server, and the work state information that the present embodiment is terminal device for speech enhan-cement supplementary carries out exemplary illustration, comprises the steps:
Step 201, obtains the voice messaging from terminal device, carries the work state information of terminal device in voice messaging.
Step 202, if judge to know that voice messaging needs to carry out speech enhan-cement process according to the device identification of terminal device, then current according to work state information determination terminal device duty, duty comprises normal operating conditions, hands-free mode duty and map mode duty.
Step 203, in multiple voice enhancement algorithms of this locality, obtains the voice enhancement algorithm corresponding with terminal device current operating state.
Step 204, the voice enhancement algorithm according to obtaining carries out speech enhan-cement process to voice messaging.
The detailed description of above-mentioned steps 201 with reference to above-mentioned steps 101, can be not described in detail in this.
In above-mentioned steps 202, such as, can have in existing all kinds gets in the terminal device of voice messaging, the voice messaging that intelligent remote controller and in-car TV receive does not need to carry out speech enhan-cement process, the voice messaging of intelligent television and panel computer needs to carry out speech enhan-cement process, therefore can carry out identification terminal equipment by the device identification of terminal device and belong to the terminal device that the terminal device needing to carry out speech enhan-cement still needs to carry out speech enhan-cement.
In one embodiment, the duty of terminal device comprises: hands-free mode duty, map mode duty, normal mode duty.Such as, under hands-free mode duty, because the microphone on terminal device can receive the voice of Correspondent Node, therefore the voice of Correspondent Node can form noise to the phonetic entry of the microphone of terminal device, under map mode duty, due to the voice message in the navigational system of map, form noise also can to the phonetic entry of microphone, and under normal mode duty, the user of terminal device is by closely sounding near microphone, and the noise in the external world can not form too large noise to the phonetic entry of microphone.
In step 203, in one embodiment, local multiple voice enhancement algorithms can comprise: the voice enhancement algorithm based on spectrum subtraction, the voice enhancement algorithm based on wavelet analysis, the voice enhancement algorithm based on Kalman filtering, the enhancing algorithm based on signal subspace, the voice enhancement algorithm based on auditory masking effect, the voice enhancement algorithm based on independent component analysis, the voice enhancement algorithm based on neural network, the voice enhancement algorithm based on deep neural network (Deep Neural Networks, DNN) etc.Correspondingly, grade classification can be carried out according to complexity to above-mentioned voice enhancement algorithm, such as, the voice enhancement algorithm of base DNN is divided into the voice enhancement algorithm of the first computation complexity, by the voice enhancement algorithm based on auditory masking effect, based on the voice enhancement algorithm of independent component analysis, voice enhancement algorithm based on neural network is divided into the voice enhancement algorithm of the second computation complexity, by the voice enhancement algorithm based on spectrum subtraction, based on the voice enhancement algorithm of Kalman filtering, Enhancement Method based on signal subspace is divided into the voice enhancement algorithm of the 3rd computation complexity.It will be understood by those skilled in the art that, more grade classification can be carried out to different voice enhancement algorithms according to computation complexity, above-mentioned first computation complexity, second computation complexity, 3rd computation complexity is only the exemplary illustration of the embodiment of the present invention, it can not form the restriction to the embodiment of the present invention, can be found out by the above-mentioned voice enhancement algorithm exemplified, the embodiment of the present invention is by performing the very high DNN voice enhancement algorithm of computation complexity on the server, can also realize carrying out a large amount of training on the server, thus obtain better speech enhan-cement model.
In step 204, such as, if terminal device is in map mode duty or hands-free mode duty, the voice enhancement algorithm of first computation complexity that computation complexity can be adopted higher carries out speech enhan-cement, if terminal device is in normal mode duty, the voice enhancement algorithm of second computation complexity that computation complexity can be adopted lower and the 3rd computation complexity carries out speech enhan-cement, make the speech enhan-cement processing procedure of terminal device to adopt thus and have more voice enhancement algorithm targetedly, the computation burden that server is unnecessary is reduced when guaranteeing speech enhan-cement quality.
As can be seen from step 201-step 204, in the present embodiment, the duty current by terminal device adopts the voice enhancement algorithm of different computation complexities to carry out speech enhan-cement to voice messaging, making speech enhan-cement to adopt thus and have more voice enhancement algorithm targetedly, reducing the computation burden that server is unnecessary when guaranteeing speech enhan-cement quality.
Refer to Fig. 3, Fig. 3 is the process flow diagram of speech enhan-cement disposal route in another way of example of the present invention, the present embodiment can be applied on server, the present embodiment carries out exemplary illustration for the ambient parameter information of speech enhan-cement supplementary environment residing for terminal device, comprises the steps:
Step 301, obtains the voice messaging from terminal device, carries the ambient parameter information of environment residing for terminal device in described voice messaging.
Step 302, if judge to know that voice messaging needs to carry out speech enhan-cement process according to the device identification of terminal device, then environmentally parameter information determination noise type.
Step 303, in multiple voice enhancement algorithms of this locality, obtains the voice enhancement algorithm corresponding with noise type.
Step 304, the voice enhancement algorithm according to obtaining carries out speech enhan-cement process to voice messaging.
The detailed description of step 301 with reference to the detailed description of above-mentioned steps 101, can be not described in detail in this.
In step 302, in one embodiment, because noise is larger by the impact of environment, therefore can be classified by environment residing for terminal device, thus can realize adopting corresponding voice enhancement algorithm to different noises, thus make to have more the process of noise reduction enhancing targetedly to the voice of terminal device.If terminal device is in (noise source is based on the noisy sound of people) in market, the voice enhancement algorithm that can adopt the sound (can be identified by frequency) eliminating people carries out speech enhan-cement, as terminal device is on road (noise source is based on the sound of blowing a whistle of vehicle), the voice enhancement algorithm that can adopt to eliminate vehicle sounds carries out speech enhan-cement, if terminal device is in classroom, better simply common voice enhancement algorithm can be adopted to carry out speech enhan-cement, make speech enhan-cement process adopt thus and have more voice enhancement algorithm targetedly.
In step 303, corresponding with the description of above-mentioned steps 302, if the ambient parameter information detected from voice messaging represents that terminal device is on road, the noise then can blown a whistle for vehicle carries out the voice enhancement algorithm of speech enhan-cement process, if the ambient parameter information detected from voice messaging represents that terminal device is in market, the voice enhancement algorithm of speech enhan-cement process then can be carried out for the excess noise of people, if the ambient parameter information detected from voice messaging represents that terminal device is in classroom, then can adopt the voice enhancement algorithm that computation complexity is lower, owing to now having more speech enhan-cement process targetedly to the voice messaging of terminal device.
As can be seen from step 301-step 304, the present embodiment is by classifying to noise source, thus voice enhancement algorithm targetedly can be had more to the voice messaging employing of terminal device, thus speech enhan-cement process targetedly can be carried out to the voice of terminal device, the voice less to noise are avoided to carry out the high voice enhancement algorithm of complexity, thus the computation complexity of speech enhan-cement can be reduced, the situation larger voice of noise being caused to speech enhan-cement poor effect owing to having carried out the lower voice enhancement algorithm of complexity can also be avoided, thus guarantee normally carrying out of later stage speech recognition.
Corresponding to above-mentioned speech enhan-cement disposal route, the application also proposed the structural drawing of the speech enhan-cement server shown in Fig. 4.Please refer to Fig. 4, at hardware view, this speech enhan-cement server comprises processor, external interface, internal memory and nonvolatile memory, certainly also may comprise the hardware required for other business.Processor reads corresponding computer program and then runs in internal memory from nonvolatile memory, and logic level is formed speech enhan-cement treating apparatus.Certainly, except software realization mode, the application does not get rid of other implementations, mode of such as logical device or software and hardware combining etc., that is the executive agent of following treatment scheme is not limited to each logical block, also can be hardware or logical device.
In order to more clearly understand the technical scheme of the embodiment of the present invention, refer to Fig. 5, Fig. 5 is the system construction drawing of speech enhan-cement process in one embodiment of the present invention, the analog voice that user is inputted by microphone is converted to digital signal after being processed by voice input module 51 by terminal device 50, after the voice compression module 56 in host CPU 52 carries out compress speech, be sent to speech enhan-cement server 53, after speech enhan-cement server 53 adopts the speech enhan-cement disposal route described in the embodiment of the present invention, voice letter after strengthening is sent to speech recognition server 54, speech recognition is carried out for speech recognition server 54, semantic understanding, the process such as phonetic synthesis, after the voice messaging of speech recognition server 54 couples of users identifies, return to speech recognition server 54 to terminal device 50 and carry out mutual voice with user, such as, after speech recognition server 54 identifies according to the voice of user, make corresponding reply, and carry out interactive voice by voice interaction module 55 with user.Wherein, voice interaction module 55 can be arranged on terminal device 50 by the mode of speech application (app), user by the current ambient parameter information of the setting options determination terminal device of app, thus can be determined at voice enhancement algorithm corresponding to current residing environment facies.Can be found out by this system construction drawing, the embodiment of the present invention can adopt the voice enhancement algorithm that complexity is higher to carry out speech enhan-cement to the voice of terminal device 50 by speech enhan-cement server 54, saves the computational resource that terminal device 50 pairs of voice carry out speech enhan-cement; In addition, can be known by above-described embodiment, directly can upgrade at speech enhan-cement server 54 pairs of voice enhancement algorithms, avoid and upgrade at terminal device 50 pairs of voice enhancement algorithms, thus improve experience when user carries out speech enhan-cement.
Please refer to Fig. 6, Fig. 6 is the building-block of logic of speech enhan-cement treating apparatus in one embodiment of the present invention, can apply on the server, and this speech enhan-cement treating apparatus can comprise:
First acquisition module 61, for obtaining the voice messaging from terminal device, carries speech enhan-cement supplementary in voice messaging;
Second acquisition module 62, if for judging to know that the voice messaging that the first acquisition module 61 obtains needs to carry out speech enhan-cement process according to the device identification of terminal device, then according to speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm;
Speech enhan-cement module 63, carries out speech enhan-cement process for the voice enhancement algorithm obtained according to the second acquisition module 62 to voice messaging.
The present invention is by the voice enhancement algorithm of the second acquisition module 62 to the voice messaging determination speech enhan-cement of the terminal device that the first acquisition module 61 gets, speech enhan-cement module 63 carries out speech enhan-cement process by corresponding voice enhancement algorithm to voice messaging, therefore avoid and carry out speech enhan-cement on the terminal device, thus alleviate the computation burden of terminal device; Because speech enhan-cement module 63 can adopt corresponding voice enhancement algorithm to the voice messaging of terminal device, thus can adopt and have more voice enhancement algorithm targetedly speech enhan-cement process is carried out to the voice messaging of terminal device, server is avoided to adopt the high voice enhancement algorithm of computation complexity to carry out unnecessary speech enhan-cement process to the voice messaging of terminal device, reduce server computation complexity when carrying out speech enhan-cement process substantially, improve the quality of speech enhan-cement, and then make follow-up speech recognition more accurate.
Please refer to Fig. 7, Fig. 7 is the building-block of logic of speech enhan-cement treating apparatus in another embodiment of the present invention, and the present embodiment is described on the basis of above-mentioned Fig. 6 embodiment.
In one embodiment, speech enhan-cement supplementary can be the work state information of terminal device, and the second acquisition module 62 can comprise:
First determining unit 621, for the duty current according to work state information determination terminal device, duty comprises normal operating conditions, hands-free mode duty and map mode duty;
First acquiring unit 622, in multiple voice enhancement algorithms of this locality, obtains the voice enhancement algorithm corresponding with the terminal device current operating state that the first determining unit 621 is determined.
First acquiring unit 622 obtains by the duty that terminal device is current the voice enhancement algorithm adopted voice messaging, making speech enhan-cement to adopt thus and have more voice enhancement algorithm targetedly, reducing the computation burden that server is unnecessary when guaranteeing speech enhan-cement quality.
In another embodiment, the ambient parameter information of speech enhan-cement supplementary environment residing for terminal device, the second acquisition module 62 can comprise:
Second determining unit 623, for environmentally parameter information determination noise type;
Second acquisition unit 624, in multiple voice enhancement algorithms of this locality, obtains the voice enhancement algorithm corresponding with the noise type that the second determining unit 623 is determined.
Second acquisition unit 624 obtains by the ambient parameter information that terminal device is current the voice enhancement algorithm adopted voice messaging, making speech enhan-cement to adopt thus and have more voice enhancement algorithm targetedly, reducing the computation burden that server is unnecessary when guaranteeing speech enhan-cement quality.
By describing above and can finding out, speech enhan-cement disposal route provided by the invention and device, according to speech enhan-cement supplementary in multiple voice enhancement algorithms of this locality, obtain corresponding voice enhancement algorithm, owing to being suitable on the server, therefore avoid and carry out speech enhan-cement on the terminal device, thus alleviate the computation burden of terminal device; Due to corresponding voice enhancement algorithm can be adopted to the voice messaging of terminal device, avoid unnecessary voice enhancement algorithm and enhancing process is carried out to voice messaging, make that speech enhan-cement is carried out to voice messaging and have more specific aim, thus server computation complexity when carrying out speech enhan-cement process can be reduced substantially, improve the quality of speech enhan-cement, and then make follow-up speech recognition more accurate.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410834628.1A CN104575509A (en) | 2014-12-29 | 2014-12-29 | Voice enhancement processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410834628.1A CN104575509A (en) | 2014-12-29 | 2014-12-29 | Voice enhancement processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104575509A true CN104575509A (en) | 2015-04-29 |
Family
ID=53091409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410834628.1A CN104575509A (en) | 2014-12-29 | 2014-12-29 | Voice enhancement processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104575509A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966517A (en) * | 2015-06-02 | 2015-10-07 | 华为技术有限公司 | Voice frequency signal enhancement method and device |
| CN108231086A (en) * | 2017-12-24 | 2018-06-29 | 航天恒星科技有限公司 | A kind of deep learning voice enhancer and method based on FPGA |
| CN108873987A (en) * | 2018-06-02 | 2018-11-23 | 熊冠 | A kind of intelligence control system and method for stereo of stage |
| CN109087659A (en) * | 2018-08-03 | 2018-12-25 | 三星电子（中国）研发中心 | Audio optimization method and apparatus |
| CN110085223A (en) * | 2019-04-02 | 2019-08-02 | 北京云知声信息技术有限公司 | A kind of voice interactive method of cloud interaction |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5570423A (en) * | 1994-08-25 | 1996-10-29 | Alcatel N.V. | Method of providing adaptive echo cancellation |
| CN1329453A (en) * | 2000-06-21 | 2002-01-02 | 阿尔卡塔尔公司 | Telephony and hand-free speed of wireless terminal equipment with echo compensation |
| US20050182624A1 (en) * | 2004-02-16 | 2005-08-18 | Microsoft Corporation | Method and apparatus for constructing a speech filter using estimates of clean speech and noise |
| CN1875611A (en) * | 2003-11-20 | 2006-12-06 | 摩托罗拉公司(在特拉华州注册的公司) | Method and apparatus for adaptive echo and noise control |
| CN101583996A (en) * | 2006-12-30 | 2009-11-18 | 摩托罗拉公司 | A method and noise suppression circuit incorporating a plurality of noise suppression techniques |
| CN102014205A (en) * | 2010-11-19 | 2011-04-13 | 中兴通讯股份有限公司 | Method and device for treating voice call quality |
| CN102801861A (en) * | 2012-08-07 | 2012-11-28 | 歌尔声学股份有限公司 | Voice enhancing method and device applied to cell phone |
| CN103456305A (en) * | 2013-09-16 | 2013-12-18 | 东莞宇龙通信科技有限公司 | Terminal and speech processing method based on multiple sound collecting units |
| CN103489452A (en) * | 2013-09-24 | 2014-01-01 | 小米科技有限责任公司 | Method and device for eliminating call noise and terminal device |
| CN104036786A (en) * | 2014-06-25 | 2014-09-10 | 青岛海信信芯科技有限公司 | Method and device for denoising voice |
| CN104052886A (en) * | 2014-06-27 | 2014-09-17 | 联想(北京)有限公司 | Information processing method and electronic device |
| CN104092801A (en) * | 2014-05-22 | 2014-10-08 | 中兴通讯股份有限公司 | Intelligent terminal call noise reduction method and intelligent terminal |
-
2014
- 2014-12-29 CN CN201410834628.1A patent/CN104575509A/en not_active Application Discontinuation
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5570423A (en) * | 1994-08-25 | 1996-10-29 | Alcatel N.V. | Method of providing adaptive echo cancellation |
| CN1329453A (en) * | 2000-06-21 | 2002-01-02 | 阿尔卡塔尔公司 | Telephony and hand-free speed of wireless terminal equipment with echo compensation |
| CN1875611A (en) * | 2003-11-20 | 2006-12-06 | 摩托罗拉公司(在特拉华州注册的公司) | Method and apparatus for adaptive echo and noise control |
| US20050182624A1 (en) * | 2004-02-16 | 2005-08-18 | Microsoft Corporation | Method and apparatus for constructing a speech filter using estimates of clean speech and noise |
| CN101583996A (en) * | 2006-12-30 | 2009-11-18 | 摩托罗拉公司 | A method and noise suppression circuit incorporating a plurality of noise suppression techniques |
| CN102014205A (en) * | 2010-11-19 | 2011-04-13 | 中兴通讯股份有限公司 | Method and device for treating voice call quality |
| CN102801861A (en) * | 2012-08-07 | 2012-11-28 | 歌尔声学股份有限公司 | Voice enhancing method and device applied to cell phone |
| CN103456305A (en) * | 2013-09-16 | 2013-12-18 | 东莞宇龙通信科技有限公司 | Terminal and speech processing method based on multiple sound collecting units |
| CN103489452A (en) * | 2013-09-24 | 2014-01-01 | 小米科技有限责任公司 | Method and device for eliminating call noise and terminal device |
| CN104092801A (en) * | 2014-05-22 | 2014-10-08 | 中兴通讯股份有限公司 | Intelligent terminal call noise reduction method and intelligent terminal |
| CN104036786A (en) * | 2014-06-25 | 2014-09-10 | 青岛海信信芯科技有限公司 | Method and device for denoising voice |
| CN104052886A (en) * | 2014-06-27 | 2014-09-17 | 联想(北京)有限公司 | Information processing method and electronic device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966517A (en) * | 2015-06-02 | 2015-10-07 | 华为技术有限公司 | Voice frequency signal enhancement method and device |
| CN104966517B (en) * | 2015-06-02 | 2019-02-01 | 华为技术有限公司 | A kind of audio signal Enhancement Method and device |
| CN108231086A (en) * | 2017-12-24 | 2018-06-29 | 航天恒星科技有限公司 | A kind of deep learning voice enhancer and method based on FPGA |
| CN108873987A (en) * | 2018-06-02 | 2018-11-23 | 熊冠 | A kind of intelligence control system and method for stereo of stage |
| CN109087659A (en) * | 2018-08-03 | 2018-12-25 | 三星电子（中国）研发中心 | Audio optimization method and apparatus |
| CN110085223A (en) * | 2019-04-02 | 2019-08-02 | 北京云知声信息技术有限公司 | A kind of voice interactive method of cloud interaction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9940935B2 (en) | Method and device for voiceprint recognition | |
| TWI582753B (en) | Method, system, and computer-readable storage medium for operating a virtual assistant | |
| US9685161B2 (en) | Method for updating voiceprint feature model and terminal | |
| JP6393730B2 (en) | Voice identification method and apparatus | |
| US10204626B2 (en) | Method and apparatus for recognizing speech by lip reading | |
| US10217463B2 (en) | Hybridized client-server speech recognition | |
| KR102103057B1 (en) | Voice trigger for a digital assistant | |
| CN103634472B (en) | User mood and the method for personality, system and mobile phone is judged according to call voice | |
| TWI684148B (en) | Grouping processing method and device of contact person | |
| KR20190100334A (en) | Contextual Hotwords | |
| CN103065631B (en) | A kind of method of speech recognition, device | |
| US10117032B2 (en) | Hearing aid system, method, and recording medium | |
| CN107799126B (en) | Voice endpoint detection method and device based on supervised machine learning | |
| CN103650035B (en) | Via social graph, speech model and the user context identification people close to mobile device users | |
| CN105940407B (en) | System and method for assessing the intensity of audio password | |
| US20160162469A1 (en) | Dynamic Local ASR Vocabulary | |
| US20180336888A1 (en) | Method and Apparatus of Training Acoustic Feature Extracting Model, Device and Computer Storage Medium | |
| WO2016209444A1 (en) | Language model modification for local speech recognition systems using remote sources | |
| JP5644013B2 (en) | Speech processing | |
| CN107147618B (en) | User registration method and device and electronic equipment | |
| EP2987312B1 (en) | System and method for acoustic echo cancellation | |
| JP5928606B2 (en) | Vehicle-based determination of passenger's audiovisual input | |
| US20150120291A1 (en) | Scene Recognition Method, Device and Mobile Terminal Based on Ambient Sound | |
| US20150095027A1 (en) | Key phrase detection | |
| US20130195285A1 (en) | Zone based presence determination via voiceprint location awareness |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| C06 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| C10 | Entry into substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150429 |
|
| RJ01 | Rejection of invention patent application after publication |